dungba88 commented on code in PR #12624:
URL: https://github.com/apache/lucene/pull/12624#discussion_r1395069004
##
lucene/core/src/java/org/apache/lucene/util/fst/BytesStore.java:
##
@@ -337,11 +349,23 @@ public long size() {
return getPosition();
}
+ /** Similar to
dungba88 commented on code in PR #12624:
URL: https://github.com/apache/lucene/pull/12624#discussion_r1393462969
##
lucene/core/src/java/org/apache/lucene/util/fst/OnHeapFSTStore.java:
##
@@ -64,22 +66,13 @@ public FSTStore init(DataInput in, long numBytes) throws
IOException {
dungba88 commented on PR #12624:
URL: https://github.com/apache/lucene/pull/12624#issuecomment-1813856323
Seems like this PR is getting long, so I spawned 2 PR out of it:
- https://github.com/apache/lucene/pull/12814: Simplify `BytesStore`
operations (which was changed to GrowableByteArra
MarcusSorealheis commented on PR #12626:
URL: https://github.com/apache/lucene/pull/12626#issuecomment-1813833061
Looks good now.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific com
dungba88 commented on code in PR #12624:
URL: https://github.com/apache/lucene/pull/12624#discussion_r1395069004
##
lucene/core/src/java/org/apache/lucene/util/fst/BytesStore.java:
##
@@ -337,11 +349,23 @@ public long size() {
return getPosition();
}
+ /** Similar to
dungba88 commented on code in PR #12624:
URL: https://github.com/apache/lucene/pull/12624#discussion_r1393547261
##
lucene/core/src/java/org/apache/lucene/util/fst/BytesStore.java:
##
@@ -21,12 +21,13 @@
import java.util.List;
import org.apache.lucene.store.DataInput;
import
dungba88 commented on code in PR #12624:
URL: https://github.com/apache/lucene/pull/12624#discussion_r1395069004
##
lucene/core/src/java/org/apache/lucene/util/fst/BytesStore.java:
##
@@ -337,11 +349,23 @@ public long size() {
return getPosition();
}
+ /** Similar to
dungba88 commented on code in PR #12624:
URL: https://github.com/apache/lucene/pull/12624#discussion_r1395069004
##
lucene/core/src/java/org/apache/lucene/util/fst/BytesStore.java:
##
@@ -337,11 +349,23 @@ public long size() {
return getPosition();
}
+ /** Similar to
Shibi-bala commented on PR #12626:
URL: https://github.com/apache/lucene/pull/12626#issuecomment-1813343695
@uschindler Ah I needed to re-sync my forked repo 😅
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL a
stefanvodita commented on issue #12796:
URL: https://github.com/apache/lucene/issues/12796#issuecomment-1813265420
+1 to starting super simple. I tried to hack a workflow for marking stale
PRs (#12813). Fortunately, GitHub provides good
[support](https://github.com/actions/stale) for this t
stefanvodita opened a new pull request, #12813:
URL: https://github.com/apache/lucene/pull/12813
PRs get stale and we miss out on good contributions. This workflow will mark
PRs that are becoming stale.
Addresses #12796
--
This is an automated message from the Apache Git Service.
benwtrent merged PR #12811:
URL: https://github.com/apache/lucene/pull/12811
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscr...@lucene.a
benwtrent opened a new pull request, #12811:
URL: https://github.com/apache/lucene/pull/12811
the `quantile` parameter is actually a `confidence_interval` this is a
simple rename of this parameter for the hnsw scalar quantized format.
--
This is an automated message from the Apache Git Se
chadbrewbaker commented on issue #12502:
URL: https://github.com/apache/lucene/issues/12502#issuecomment-1813112623
> Yes:
>
> * no external libraries for Lucene Core
> * no native code
Put it in an "examples" directory to show how to extend Lucene with JNI. If
you have a $1
jpountz commented on PR #12789:
URL: https://github.com/apache/lucene/pull/12789#issuecomment-1813030726
++ This feels similar to `IndexOrDocValuesQuery`: we probably can't guess
the absolute best threshold, but we can probably figure out something that is
right more often than wrong. Hopef
benwtrent merged PR #12806:
URL: https://github.com/apache/lucene/pull/12806
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscr...@lucene.a
kaivalnp commented on PR #12679:
URL: https://github.com/apache/lucene/pull/12679#issuecomment-1812956627
> could you test on cohere with Max-inner product?
Thanks, the gist was really helpful and gave some files including normalized
and un-normalized vectors. I assume that since you
kaivalnp commented on PR #12679:
URL: https://github.com/apache/lucene/pull/12679#issuecomment-1812941899
> You still need to score the vectors to realize that they are in the
iteration set or not
Right, I meant that we need not score all *other* vectors to determine if
the vector it
benwtrent commented on issue #12627:
URL: https://github.com/apache/lucene/issues/12627#issuecomment-1812901631
@nitirajrathore could you add something to
[KnnGraphTester](https://github.com/mikemccand/luceneutil/blob/master/src/main/KnnGraphTester.java)
that is a test for connectedness?
msokolov commented on PR #12810:
URL: https://github.com/apache/lucene/pull/12810#issuecomment-1812678173
this sounds reasonable to me, and the code does seem simpler, but I'm not
able to give a thorough review. +1 to rationalize / simplify even if it doesn't
show significant peformance imp
msokolov commented on issue #12627:
URL: https://github.com/apache/lucene/issues/12627#issuecomment-1812647124
My memory of the way this diversity criterion has evolved is kind of hazy,
but I believe in the very first implementation we would not impose any
diversity check until the neighbor
jpountz commented on code in PR #12806:
URL: https://github.com/apache/lucene/pull/12806#discussion_r1394256922
##
lucene/core/src/java/org/apache/lucene/codecs/lucene99/Lucene99HnswVectorsReader.java:
##
@@ -238,11 +238,23 @@ public void search(String field, float[] target,
Kn
benwtrent commented on issue #12627:
URL: https://github.com/apache/lucene/issues/12627#issuecomment-1812605864
@msokolov good point.
It seems to me we would only fully disconnect a sub-graph only if its very
clustered. Is there a way to detect this in the diversity selection?
msokolov commented on issue #12627:
URL: https://github.com/apache/lucene/issues/12627#issuecomment-1812559330
Is the problem primarily to do with single isolated nodes or do we also see
disconnected subgraphs containing multiple nodes? I think this idea would
prevent the isolated nodes, bu
mikemccand commented on PR #12791:
URL: https://github.com/apache/lucene/pull/12791#issuecomment-1812469386
Nightly sparse (NYC taxis) benchy was a bit unhappy with this change because
it (weirdly) relies on `Query.toString` (I tried to fix the benchy
[here](https://github.com/mikemccand/lu
benwtrent commented on PR #12806:
URL: https://github.com/apache/lucene/pull/12806#issuecomment-1812421692
> The idea makes sense to me, what is less clear to me is whether this logic
belongs to the Query or to the vector reader: should searchNearestNeighbors
implicitly do a linear scan whe
benwtrent commented on issue #12627:
URL: https://github.com/apache/lucene/issues/12627#issuecomment-1812408858
@nitirajrathore @msokolov I had an idea around this, and it will cost an
extra 4bytes per node on each layer its a member (maybe we only need this on
the bottom layer...)
W
jpountz commented on code in PR #12806:
URL: https://github.com/apache/lucene/pull/12806#discussion_r1394090752
##
lucene/core/src/java/org/apache/lucene/search/AbstractKnnVectorQuery.java:
##
@@ -110,6 +110,12 @@ private TopDocs getLeafResults(LeafReaderContext ctx,
Weight fil
jpountz commented on PR #12810:
URL: https://github.com/apache/lucene/pull/12810#issuecomment-1812099057
For reference, starting postings and skip lists at -1 changes file formats,
so I'm keen to getting this change in 9.9 since we had to change the file
format anyway because of the move fr
jpountz commented on PR #12810:
URL: https://github.com/apache/lucene/pull/12810#issuecomment-1812097551
This change seems to be neutral on wikibigall. No speedup, but not slowdown
either.
```
TaskQPS baseline StdDevQPS
my_modified_version Std
dungba88 commented on code in PR #12624:
URL: https://github.com/apache/lucene/pull/12624#discussion_r1393547261
##
lucene/core/src/java/org/apache/lucene/util/fst/BytesStore.java:
##
@@ -21,12 +21,13 @@
import java.util.List;
import org.apache.lucene.store.DataInput;
import
jpountz opened a new pull request, #12810:
URL: https://github.com/apache/lucene/pull/12810
Currently `advance(int target)` needs to perform two checks:
- is there a need to use skip lists?
- is there a need for decoding a new block?
Ideally we would track the last doc ID in a
picimako opened a new issue, #12809:
URL: https://github.com/apache/lucene/issues/12809
### Description
Hi,
I've been looking into how
[`org.apache.lucene.luke.app.desktop.util.TextAreaPrintStream`](https://github.com/apache/lucene/blob/main/lucene/luke/src/java/org/apache/luce
zacharymorn commented on PR #240:
URL: https://github.com/apache/lucene/pull/240#issuecomment-1811972923
> Hi @mikemccand @jpountz @javanna @gsmiller , I have updated this PR to
pick up the latest from `main`, as well as revert some changes to save them for
follow-up PRs that address other
34 matches
Mail list logo
